Simple benchmark to test the most popular open source and commercial LLMs with automated OpenCode
3 files for Claude Code, Codex and OpenCode: benchmark-audit, llm-coding-benchmark CLAUDE.md, llm-coding-benchmark AGENTS.md — 7,956 tokens loaded in every session.
CLAUDE.md E 7,516 tok AGENTS.md A 440 tok .agents/skills/benchmark-audit/SKILL.md A 74 tok These files are akitaonrails/llm-coding-benchmark's own configuration — they tell Claude Code, Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.